Update frontend.py

frontend.py

@@ -63,32 +63,33 @@ with st.sidebar:
     voice_lang = st.selectbox("🔈 Voice Language", ["en", "fr", "es", "de", "hi", "zh"])
 
 # Text-to-Speech
-def speak(text, lang='en'):
-    try:
-        speech_key = st.secrets["AZURE_SPEECH_KEY"]
-        region = st.secrets["AZURE_REGION"]
-
-        speech_config = speechsdk.SpeechConfig(subscription=speech_key, region=region)
-        audio_config = speechsdk.audio.AudioOutputConfig(use_default_speaker=False)
-
-        speech_config.speech_synthesis_voice_name = f"{lang}-US-JennyNeural" if lang == "en" else f"{lang}-Standard-A"
+# Setup usage tracking
+if "tts_usage_count" not in st.session_state:
+    st.session_state.tts_usage_count = 0
+if "enable_audio" not in st.session_state:
+    st.session_state.enable_audio = False
 
-        synthesizer = speechsdk.SpeechSynthesizer(speech_config=speech_config, audio_config=None)
-
-        result = synthesizer.speak_text_async(text).get()
-        if result.reason == speechsdk.ResultReason.SynthesizingAudioCompleted:
-            audio_data = result.audio_data
-            mp3 = BytesIO(audio_data)
-            b64 = base64.b64encode(mp3.getvalue()).decode()
-            st.markdown(f'<audio controls><source src="data:audio/mp3;base64,{b64}" type="audio/mp3"></audio>', unsafe_allow_html=True)
-            mp3.seek(0)
-            return mp3
-        else:
-            st.warning("Azure TTS failed.")
-            return BytesIO()
+# Azure TTS function
+def azure_speak(text, lang='en-US'):
+    try:
+        speech_config = speechsdk.SpeechConfig(
+            subscription=st.secrets["AZURE_SPEECH_KEY"],
+            region=st.secrets["AZURE_REGION"]
+        )
+        speech_config.speech_synthesis_language = lang
+        with tempfile.NamedTemporaryFile(delete=False, suffix=".mp3") as tmpfile:
+            audio_config = speechsdk.audio.AudioOutputConfig(filename=tmpfile.name)
+            synthesizer = speechsdk.SpeechSynthesizer(speech_config, audio_config)
+            result = synthesizer.speak_text_async(text).get()
+            if result.reason == speechsdk.ResultReason.SynthesizingAudioCompleted:
+                st.session_state.tts_usage_count += 1
+                return tmpfile.name
+            else:
+                st.error("Speech synthesis failed.")
+                return None
     except Exception as e:
-        st.error(f"🔊 Azure TTS Error: {e}")
-        return BytesIO()
+        st.error(f"Azure TTS error: {e}")
+        return None
 
 
 tab1, tab2 = st.tabs(["🧠 Analyze Review", "📚 Bulk Reviews"])
@@ -168,9 +169,20 @@ with tab1:
         except Exception as e:
             st.warning(f"🧪 Logging failed: {e}")
 
+        # Only show toggle and button for audio
         st.subheader("🔊 Audio Summary")
-        audio = speak(data["summary"], lang=voice_lang)
-        st.download_button("⬇️ Download Audio", audio.read(), "summary.mp3")
+        st.session_state.enable_audio = st.toggle("🎧 Generate Audio Summary")
+
+        if st.session_state.enable_audio:
+            if st.session_state.tts_usage_count > 20:
+                st.warning("🔇 Azure TTS usage limit reached for this session.")
+            else:
+                if st.button("▶️ Generate & Play Audio"):
+                    audio_path = azure_speak(data["summary"], lang=f"{voice_lang}-US")
+                    if audio_path:
+                        audio_bytes = open(audio_path, "rb").read()
+                        st.audio(audio_bytes, format="audio/mp3")
+                        st.download_button("⬇️ Download Audio", audio_bytes, "summary.mp3")
 
         st.markdown("### 🔁 Ask a Follow-Up")
         sentiment = data["sentiment"]["label"].lower()